Examining Teacher Expectationsabout Physics Homework
Physics Honors ThesisBy: Heather Demarest
Advisor: Noah FinkelsteinHonors Council Member: John Cumalat
Outside Member: Valerie OteroDedicated to Tony Barker
April 8th, 2004
1
Abstract
There are many different ways by which students learn physics and develop
beliefs about physics. These range from exams to lectures, from labs to homework.
Teachers have beliefs about the ideal content for each of these media to contain, as well
as beliefs about what they typically do contain. The purpose of my thesis, therefore, is to
examine in detail, a small but vital way that this information is conveyed from teacher to
student: Homework. First, I design a survey to be administered to teachers of
introductory university classes. This survey is designed to acquire data about teachers’
expectations and beliefs about their homework content. Next, I administer the survey and
simultaneously conduct an interview with each professor in my study. Then, I acquire
homework sets from the teachers’ classes. I rate these homework sets along the same
dimensions the teachers were asked to rate them. Finally, I compare the ratings and
analyze them for agreement.
2
Table of Contents
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
II. Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
a. Goals and Motivation for PER. . . . . . . . . . . . . . . . . . 5
b. Physics Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
c. Physics Expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
III. My Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
a. Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
b. Methods and Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
c. Data: Teacher Surveys and Interviews . . . . . . . . . . 25
d. Analysis: Teacher Surveys and Interviews. . . . . . . 28
e. Data: Homework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
f. Analysis: Homework . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
g. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34
IV. Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
V. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43
VI. Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
VII. Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3
I. Introduction
It may come as a surprise to many teachers that despite (or perhaps even because
of) their best efforts, students in introductory physics classes do not typically master basic
physics concepts. Additionally, these students do not acquire their professors’ attitudes
and beliefs about physics itself as a discipline. Since the primary goals of most teachers
are for their students to learn physics content as well as general beliefs about the nature of
physics, something is going wrong. As a response to this problem, a new field has arisen
within the physics community: physics education research or PER. In one particularly
prominent area of PER, researchers attempt to discover the extent to which students
succeed or fail to learn physics and their physics expectations. They also hope to uncover
why students succeed or fail. Some work on developing successful teaching methods for
a variety of subjects, students and environments.
There are many different ways by which students learn physics and develop
beliefs about physics. These range from exams to lectures, from labs to homework.
Teachers have beliefs about the ideal content for each of these media to contain, as well
as beliefs about what they typically do contain. The purpose of my thesis, therefore, is to
examine in detail, a small but vital way that this information is conveyed from teacher to
student: Homework. First, I design a survey to be administered to teachers of
introductory university classes. This survey is designed to acquire data about teachers’
expectations and beliefs about their homework content. Next, I administer the survey and
simultaneously conduct an interview with each professor in my study. Then, I acquire
homework sets from the teachers’ classes. I rate these homework sets along the same
4
dimensions the teachers were asked to rate them. Finally, I compare the ratings and
analyze them for agreement.
There are many different ways that students construct an understanding from all
of the different resources offered throughout the course. Even though the way that
students learn is very complicated, I have approximated the process by the following
model:
Teacher
Lectures Labs Homework Tutorials Textbook
Student
The area my research project addresses is circled in the above diagram. The
purpose of my research project is to discover some of the consistencies and discrepancies
between teacher and student expectations. Particularly, I focus within the medium of
homework and examine the coordination and discontinuity between the teacher’s ideal
beliefs about homework content and their actual homework practices.
Future work on this subject would be to correlate these findings with other areas
of learning as well as students’ expectations about homework content. Then it would be
possible to discover where the consistencies and discrepancies exist between the actual
homework content and student beliefs about homework content as well as between their
beliefs about homework content and their beliefs about physics in general. Also, some
Figure 1: Diagram of Teacher / Student interaction via media
5
very important future work would be to repeat this process of interviews, surveys and
analysis for other methods of learning, such as labs, lectures and tutorials, etc. An
interesting avenue for further work would also be to discover how the location of
consistencies and discrepancies correlates with the teachers’ knowledge of physics
education research.
II. Background
1. Goals and Motivation for Physics Education Research
The United States leads the world in many ways. It is a major economic, military
and entrepreneurial force. Therefore, it is surprising that its students perform so poorly in
science and math when compared to students around the world. When advanced high
school senior students from different countries were compared, the United States
performed significantly worse than the others1, leading many teachers, leaders and even
politicians to ask that effort and money be spent on figuring out why the U.S. performed
so low, as well as how to improve. The following chart is based on the results from a test
that high school students took in 1998. They were asked a broad range of questions,
requiring them to solve traditional-type physics problems as well as conceptual problems
about physics.
1 http://timss.bc.edu Mathematics and Science Achievement in the Final Year of Secondary School.February 1998. Mullis, Martin, Beaton, Gonzalez,Kelly, and Smith(http://timss.bc.edu/timss1995i/TOC_MSC.html)
6
These results were problematic for policy-makers in the U.S. Many emphasized
the importance of physics education, and it was in 2000 that, “The [National]
Commission [on Mathematics and Science Teaching for the 21st Century] is convinced
that the future well-being of our nation and people depends not just on how well we
educate our children generally, but on how well we educate them in mathematics and
Figure 2: Comparison of High school physics achievement between countries
7
science specifically.”2 Congress agreed and even passed a law demanding that U.S.
students perform better in the future.3
The American Physical Society further supported the field, stating, “In recent
years, physics education research has emerged as a topic of research within physics
departments . . . The APS applauds and supports the acceptance in physics departments
of research in physics education.”4
2. Physics Concepts
As teachers and researchers in physics realized that their students were not
acquiring basic underlying physics concepts, they designed tests to quantify the extent to
which their students’ knowledge was insufficient. These are termed ‘concept tests’ and
there are a lot of them. A frequently used and cited example of these is the Force
Concept Inventory, or FCI5. Through a series of multiple choice questions, it attempts to
gauge students’ understanding of the concepts most teachers agree students should have
mastered by the end of an introductory mechanics course. A sample question of the FCI
is as follows6:
2 Before It’s Too Late (Glen Commission)- September 2000.3 “By the Year 2000, United States students will be first in the world in mathematics and scienceachievement”Goals 2000: Educate America Act. Pub. Law 103-227 passed: 4/26/964 American Physical Society: Statement on Research in Physics Education (1999)5 Hestenes, Wells, and Swackhamer, “Force Concept Inventory,” Physics Teacher 30 141-158 (1992).6 Hestenes, Wells, and Swackhamer, “Force Concept Inventory.” Physics Teacher, v. 30 141-158, (1992).
8
Students who have merely mastered the equation-matching or a ‘plug and chug’
approach will not be able to answer this question because it relies on an understanding of
the underlying concepts about force. Further, this question can show when a student does
not even utilize concepts of forces to answer force-related physics questions, (for instance
when he or she chooses answer ‘C’). Answer C is a good example of a ‘research-based
distractor,’ which is a common, implicitly held belief of students. These researched-
based distractors are discovered by researchers in physics education who use interviews
observations and surveys of students to ascertain their beliefs. Often-times physics
teachers do not think to provide C as an answer, and are incredulous to learn that many
students, who can answer quantitative problems about forces, hold this view,
demonstrating that they have failed to learn even the most basic concepts about forces.
This and other concept tests have been used to measure students’ success at
Figure 3: Sample question from the Force Concept Inventory
Imagine a head-on collision between a large truck and a small compact car.
During the collision:(A) The truck exerts a greater amount of force on the car than the car exerts on
the truck.
(B) The car exerts a greater amount of force on the truck than the truck exerts
on the car.
(C) Neither exerts a force on the other; the car gets smashed simply because it
gets in the way of the truck.
(D) The truck exerts a force on the car but the car does not exert a force on thetruck.
(E) The truck exerts the same amount of force on the car as the car exerts on the
truck.
Force Concept Inventory (1992)
9
physics. The results from these tests can then be quantitatively compared between
classes, schools, and countries, as in the previously discussed TIMSS study.
The students’ poor performance on these concept tests combined with teachers’,
scientists’, and politicians’ desire to improve led to many educational reforms, especially
in lecture styles and the design of lab experiments. These interventions are based on
promoting conceptual development. One example of such a change is the use of Personal
Response Systems, or ‘clickers,’ in large lectures. These ‘clickers’ are connected via
infrared signal to a computer which can immediately tally students’ answers to a question
asked during lecture. This gives teachers instantaneous feedback about the level of
understanding of their students. They can cater their lectures in real time to the needs of
the class. Also, students are often asked to discuss their answers with their peers before
they punch in their answers. In this way, they are also participating in a form of peer
instruction.7 Another popular change was to introduce tutorials.8 A tutorial is a way of
leading students in a Socratic, reasoned way through a complex, conceptual problem. It
is usually done in addition to lecture.
These kinds of changes have resulted in higher student performance on concept
tests such as the FCI. The following graph shows the gain (how much students’
conceptual understanding improve over the course of the semester) both in traditional
classes without research based changes (black) and in interactive engagement (IE) classes
7 E. Mazur, Peer Instruction: A User's Manual (Englewood Cliffs, NJ: Prentice-Hall, 1997).http://www.columbia.edu/cu/gsapp/BT/RESEARCH/mazur.html8 McDermott, Lillian, Department of Physics, University of WA, Seattle
10
which incorporate research based changes ranging from peer instruction to tutorials to
wholesale course reform (grey).9
<g> = gain = Post-class score – Pre-class score100% – Pre-class score%
For example, suppose a student scores a 40% before he takes the class and an
80% on the same test after taking the class. His gain would be:
<g> = 80% - 40% = .66100% - 40%
Notice that most students do not score nearly as high as our example above. The
average gain for a traditional class is only .23. With the interactive engagement, the
average is close to .5. Not only is the average gain higher in classes utilizing IE, but
9 Hake, R.R. "Interactive-engagement vs. traditional methods: A six-thousand-student survey of mechanicstest data for introductory physics courses," Am. J. Phys. 66, 64-74 (1998).http://www.physics.indiana.edu/~sdi/ajpv3i.pdf
Figure 4: Comparison of gain between traditional and interactive engagement classes
14 traditional courses N=208448 IE courses N=4458
11
additionally, almost every student performs higher than almost all of the students in
traditional classes.
3. Physics Expectations
Sometimes teachers, even in reform courses, assume that if their students are
performing well on conceptual tests, their general beliefs about physics must be correct as
well. For instance, some teachers assume that students who can correctly answer
conceptual questions about circuits also believe that physics is a coherent, conceptual
field about the everyday real world. These teachers assume that beliefs about physics
‘come along for the ride’ with knowledge of conceptual physics content. In reality, many
students who can answer conceptual physics questions do not always believe that physics
is a coherent, conceptual field about the everyday world. Worse, their beliefs about
physics typically deteriorate over the course of the semester. In other words, if a student
enters a class believing that physics is about the real world, most likely, she will leave the
class with only a weak belief that physics is about the real world. One reason this
deterioration is problematic is because my data suggest that conveying these beliefs is
one of their main goals in the course. For example, in an interview I conducted, one
teacher claimed his primary goal was for students to “understand that physics explains
the world around them.”10 Teachers want their students to share their beliefs about the
intrinsic nature of physics. Another reason this deterioration is troublesome is that
preliminary data11 show that student beliefs about physics are correlated with retention.
In other words, students that believe physics is a coherent, conceptual study of the real
10 Teacher A, see appendix11 Adams, Finkelstien, Wieman, forthcoming.
12
world are more likely to go on in physics than those who do not have those beliefs. It is
important to note that this is merely a correlation, not causation.
Maryland Physics Expectations Survey
One of the most common tools used to ascertain student beliefs is a survey called
the Maryland Physics Expectations Survey (MPEX).12 The survey has 34 statements
about the nature of physics. Students are asked to choose a number, 1 through 5,
depending on whether they agree or disagree with the statement. 1 is “strongly disagree,”
2 is “disagree,” three is “neutral,” four is “agree,” and five is “strongly agree.” The
‘expert view’ was derived by giving the survey to physics teachers. The teachers’ answer
is the ‘expert’ answer. Answers of 4 and 5 are grouped together as “agree” and answers
of 1 and 2 are grouped together as “disagree.” If the student agrees with the expert view,
her answer is considered favorable. If she disagrees, her answer is considered
unfavorable. For example, one of the MPEX statements is: “Physical laws have little
relation to what I experience in the real world.” The expert answer is ‘disagree.’
(Presumably this is because physical laws do relate to what a person experiences in the
real world.) Therefore, if a student answers with a 4 or a 5, her answer is unfavorable,
(because she disagrees with the expert) whereas if she answers with a 1 or a 2 her answer
is favorable (because she agrees with the expert). The MPEX is designed to ascertain
students’ beliefs and expectations along six dimensions of physics.
The first dimension is independence. These statements ask the student if physics
should be learned independently with the student taking responsibility for his or her own 12 Redish, Saul and Steinberg, “Student Expectations in introductory physics,”Maryland PhysicsExpectations Survey University of Maryland Physics Education Research Group. American Journal ofPhysics 66 212-224, (1998). http://www.physics.umd.edu/perg/expects/mpex.htm. (See Appendix C1)
13
understanding, or if physics should be learned without evaluation by taking what the
teacher and text book say as given. The first answer is considered favorable.
The second dimension is coherence. These statements ask the student if physics
should be considered as a connected, consistent framework or if physics can be treated as
separated facts or ‘pieces’. Again, the first is the favorable answer.
The third dimension is concepts. These statements ask the student if physics
stresses the understanding of the underlying ideas and concepts or if studying physics
means focusing on memorizing and using formulas. The first response is favorable.
The fourth dimension is the reality link. These statements ask students if they
believe ideas learned in physics are relevant and useful in a wide variety of real contexts
or if they believe ideas learned in physics are unrelated to experiences outside the class-
room. The first beliefs are favorable.
The fifth dimension is math link. These statements ask if students consider
mathematics as a convenient way of representing physical phenomena or if they view the
physics and the math independently with no relationship between them. Again, the first
answer is favorable.
The sixth dimension is effort. These statements ask if the student makes the effort
to use information available and tries to make sense of it or if the student does not
attempt to use available information effectively. The first answer is the favorable one.
If a student agrees with the expert view on every single statement, then her
expectations would be entered at the top left corner of the following graph,13 at the point
(100%, 0%). If she disagrees with the expert view on every single statement, her
expectations would be entered as (0%, 100%). If she is neutral on every statement, her 13 Redish, 1998.
14
expectations would be entered as (0%, 0%) because she has neither favorable nor
unfavorable expectations. The arrows represent the average beliefs before and after the
course. With the minor exception of the ‘concepts’ category, all of the expectations
deteriorate between the beginning of the class and the end. In the category of ‘math’ they
drop by as much as 10%.
Colorado Learning Attitudes about Science Survey
Researchers at the University of Colorado, Boulder have built on the MPEX and
other instruments to develop the Colorado Learning Attitudes about Science Survey or
CLASS. It is a similar survey designed to ascertain students’ attitudes and expectations
Figure 5: Pre to Post-class changes in expectations
0 20 40 60 80 100Unfavorable
Deterioration of Expectations during asemester introductory course (MPEX)
15
about physics. Comparable results have been obtained with the CLASS that also
demonstrate students’ expectation regression during an introductory course. Results from
the CLASS can be seen in the table below.14 Data taken before the semester of
introductory physics are “Pre,” while data taken after the semester are “Post.” As can be
seen, there is a significant drop in expectations over the course of the semester. In other
words, the students agreed with almost 70% of statements asserting that physics is about
the real world before they took the course. After the course and all of its lectures, exams,
labs and homework, these same students agreed with only 52% of the same statements
asserting that physics is about the real world. The following data is from an introductory
course at the University of Northern Colorado.
Category CLASS Pre Favorable% CLASS Post Favorable%
reality (world) 69.5 (4.5%) 52reality (personal) 66 (4%) 42.5independence 61 (3%) 49coherence 58.5 (3%) 58concepts 59 (3%) 47.5math 73.5 (3.5%) 59.5effort 40 (2.5%) 32.5
As I mentioned before, preliminary data show that students whose views are
correlated with the expert views are much more likely to go on in physics15. The
following table shows the percent favorable responses to CLASS category, Reality
Personal View, (what I’ve been calling ‘Student Reality’ link) issued at the beginning of
term for a variety of different types of courses:
14 Adams, Finkelstein, Wieman, Forthcoming.15 Adams, Finkelstien, Wieman, forthcoming.
Figure 6: Sample Pre to Post-class changes in expectations for U.N.C. intro class
16
University EnvironmentFavorablePre-Test Score (%)(uncertainty; n)
UNC Calc-based Physics I (FCI 0.35)(mostly majors)
71 (5%, n=41)
CU Calc-based Physics I (FCI 0.6)(mostly engineers)
63 (2%, n=174)
UNC Alg-based Physics I (FCI 0.13)(who stayed enrolled) Students who started
61 (5%, n=36)
49 (4%, n=78)CU Non-Science Major Physics I 44 (4%, n=77)
CU Non-Science Major Physics II 61 (n=34)
In the above graph, notice that for the UNC Algebra-based Physics I class the
students who stayed enrolled and went on to complete the course had significantly higher
(61% as opposed to 49%) pre expectations. In other words, this result is not due to any
expectations they learned by staying in the course. Similarly, the students that chose to
go on to the Non-Science Major Physics II had higher expectations for how much physics
related to their lives than the general population of those entering the Non-Science Major
Physics I class.16
III. My Project
There are many ways these expectations are taught, both explicitly and implicitly.
Teachers convey these through what they test on exams, what they say in lecture, what
they ask their students to do in labs, which text books they assign and in which
homework problems they assign. In order to keep my project down to a manageable size,
I had to pick one of these areas to examine in detail.
16 Please note that this only establishes a correlation, not a causal relationship between expectations andphysics retention.
Figure 7: Expectation data for a variety of classes in one expectation dimension
17
I decided to focus on homework. My reasons for choosing homework are
threefold. First, most teachers use homework scores as a significant part of the class
grade. For example, one of the teachers I surveyed has the following statement on his
course website: With this system, the most important requirement for getting a good
grade is to do all the assignments! (Bold emphasis original.) Second, students spend a
large amount of time working on homework and teachers spend a large amount of time
finding and grading assignments homework. This time should be time well-spent.
Finally, and most importantly, many teachers believe that it is by ‘doing problems’ that
students learn about physics and how real physics is done.
1. Hypothesis
As observed in the data above, not only is there a discrepancy between expert and
student expectations about physics, but that discrepancy widens as student expectations
regress during the course. Therefore, since homework is a major part of what physics
students are required to do, it is reasonable to look for evidence of that discrepancy
somewhere within homework.
The model I use to describe how teachers convey expectations to students through
homework is a magnified subset of the model shown in the introduction:
Teacher
Homework
Student
The diagram below contains the details I am going to examine. The arrows
represent connections between the different elements and the lines through them are
possible discrepancies or breakdown points.
18
My hypothesis is that a breakdown can be observed between teachers’ ideas of
physics / ideal homework content and actual homework content. (These possible
locations are indicated by the question marks.) My thesis will investigate the extent and
characteristics of this breakdown.
In order to keep my project a manageable size, I limit it in two ways. First, my
research only considers the top three categories, between teacher expectations and
perceptions of homework and homework content. It is not within the scope of this
project to consider the students’ categories. I will suggest how this could be done at the
end of the thesis in the section on further work. Next, I will choose three out of the six
MPEX dimensions of expectations presented above to examine. Based on the CLASS, I
will divide the reality link into two categories, which I will discuss below, leaving me
with four expectation dimensions to analyze in the homework. Specifically, I will
Figure 8: Model of interaction between teachers and students within homework
Teachers’ Perception of Expectations in Homework
Teachers’ Ideas of Physics and Ideal Homework
Expectations found from Evaluation of Homework
Students’ Perception of Expectations in Homework
Students’ Ideas of Physics
19
examine coherence, concepts and two aspects of reality link (real world and student
world).
2. Methods and Tools
The first step in my project was to design a survey to ascertain teacher beliefs
about the content of their homework as well as the relative importance faculty place on
homework. I asked teachers to list the top five goals for their class. I also asked what
percentage of learning they believe occurs during a number of different activities: during
lectures, while doing homework, while reading the textbook, while preparing for exams,
during recitations or labs, or somewhere else. Using the MPEX and CLASS as guides, I
first formulated a preliminary survey17 with eleven statements concerning the expectation
dimensions. Teachers are asked to indicate whether they agree, disagree or are neutral
with respect to the statements. They are asked to give a number (just as before, 1 is
strongly disagree, 3 is neutral, and 5 is strongly agree) for each of the various statements.
Designing a survey that could accurately capture teachers’ expectations proved to be
much more difficult than I had anticipated.
One problem I had with the first survey was that many of the statements did not
present a dimension clearly enough to accurately categorize a teacher’s beliefs. For
instance, one statement reads: “The overall coherence of physics is often explicit in
homework.” Many teachers found it difficult to rate their agreement because it wasn’t
clear what would constitute “explicit coherence.” This statement was taken out of the
second version of the survey.
17 See Appendix A1 for a copy of survey #1.
20
Another problem brought to my attention by Wendy Adams18 is that the two
statements about the reality link really contain two, very different ideas. The first
statement says, “Homework problems frequently ask the student to consider a real world
situation to which a physics principle applies.” The second says, “Homework questions
often ask the student to consider his or her own experience.” Most teachers agree with
the first statement but disagree with the second. This is because, as one teacher said, “all
of physics is about the real world,” independently of whether or not students are asked to
consider their experience. To address this problem, as was done in the CLASS, I divided
the category of reality link into two categories: real world link and student world link. I
will discuss further problems with the real world link below. Another problematic
expectation dimension in the first survey attempted to gauge teachers’ expectations about
effort. This was very difficult to measure because teachers were unwilling to make
general statements about how effort affected the students’ success in their class. For
instance, one of the questions I asked on the first survey is: If students do not do well on
homework assignments it is generally because they are not putting in as much time and
effort as I have recommended. Many teachers said that this was highly student-
dependent. Some noted that a student may be putting in plenty of time and effort, but in
the ‘wrong ways’ (not in groups, or just re-reading the text book). They pointed out that
doing poorly on the homework could be attributed to a number of different factors, and so
did not feel comfortable generalizing enough to assign a number. I got similar reactions
to the other question about effort, so I decided to throw out the effort expectation
dimension and focus on the others.
18 Wendy Adams is a graduate student at CU currently doing research on student expectations with theCLASS survey.
21
In addition to the survey, I used teacher interviews as a tool for validating the
survey, evaluating teacher expectations and confirming their goals as well as the
importance they place on homework. I audio recorded the interviews and took notes so
that I could reliably use the interview data.
First, the interview is used to validate the survey. If, for example, I were to hand
out the surveys and collect them later, I would only have a number (1-5) for each
statement. I wouldn’t know that the teachers had trouble deciding what ‘explicit
coherence’ is, etc. By interviewing them and asking them to ‘think aloud’ as they fill out
the survey I obtain a lot of information about what they think the statements are getting at
as well as whether or not they agree.
Next, I use the interview to get direct quotations, such as those of Teacher B who
said, “I need to look at my homework” in order to answer the survey questions about the
content of his homework. Teacher B also said, “If they [the students] look at the CAPA19
they are being misled” about what is important.
The interviews also provide a more thorough understanding of teacher
expectations. For instance, a teacher might say, “I’m putting down a four because even
though I don’t have very many conceptual problems in my homework assignments, the
ones that are conceptual cannot be solved without a truly thorough conceptual
understanding of the underlying material.” This statement contains a lot more
information than a number. The interviews are especially useful for designing the next
version of the survey. The interviews are also useful for making sure my survey is
19 CAPAs are computer assisted physics assignments which are assigned for homework in this teacher’sclass.
22
comprehensive enough. I ask each teacher if he thinks I’ve left something crucial out of
the survey.
After an analysis of my preliminary survey data and the first set of interviews, I
begin the design of my final survey, adding and removing several statements.20
For the reasons discussed above (‘explicit coherence’ is difficult to characterize) I
remove the question about ‘explicit coherence’ in homework. One of the other
statements I use concerning coherence in the first survey is: Students are asked to infer
general principles from specific examples in the homework problem sets. Many teachers
did not know how to answer this question either. Therefore, I modify both coherence
questions for the next survey.
I ask the same first ten questions for general information about the importance of
homework based on what percentage of learning the teacher believes occurs while doing
homework. I also ask the teacher to list his top five goals for the class. The goals can be
used to help ascertain teachers’ ideas about physics generally and ideal homework
content. Then I ask the teachers to agree or disagree with twelve statements about the
importance of homework and the four expectation dimensions: coherence, concepts, real
world link, and student world link.
The Final Survey
The three statements about coherence are:
1. The majority of problems in a given homework set ask about thesame two or three ideas.2. If a student does poorly on the homework in the first unit, he or shecan still do well on the next unit’s homework by working exclusively onthe new material.3. There is usually at least one homework question in each set abouthow the current material relates to other material from the course.
20 See Appendix A2 for a copy of Survey #2.
23
Notice that the first and third of these statements are positive, while the second is
negative. In other words, if a teacher strongly believes his homework is coherent, he will
answer with a 5 for the first statement, a 1 for the second statement and a 5 for the third
statement. Therefore, when rating whether or not a teacher believes his homework is
coherent, it is necessary to exchange or reverse the values for all of the negative
statements. So the teacher described above would have a ‘coherence score’ of: 5 5 5.
Stated more precisely, an agreement of 1 (strong disagreement with the statement) for the
second statement is really an agreement of 5 with the belief that the homework is
coherent. Conversely, if a teacher has an agreement of 4 for the second statement, it
needs to be reversed to a 2 in order to evaluate the teacher’s belief about whether or not
his homework is coherent.
The two statements about concepts are:
• Most homework problems can be solved by carefully looking overthe given equations to find the right one.
• In order to solve my homework problems students must understandthe underlying concepts.
For this expectation, the first statement is negative and the second is positive in
the sense discussed above.
The two statements about the world reality link are:
• Homework problems frequently ask the student to consider a realworld situation to which a physics principle applies.
• Most of the ‘real world’ homework problems I assign are‘idealized’ or ‘abstracted’ to simplify the complexities of the realworld.
The first statement is positive, the second is negative.
The two statements about the student reality link are:
24
• Homework questions often ask the student to consider his or herown experience.
• The material I cover in my homework relates to the students’everyday lives.
These are both positive statements.
The three statements about the importance of homework are:
• All of the material (concepts, formulas, etc.) I expect my studentsto know on an exam is covered at some point on the homework.
• If a student does poorly on the homework, he or she will mostlikely do poorly overall in the class.
• If students want to know what I think it is important for them tolearn, they can look at the homework problems I assign.
These statements are all positive.
The final tool I utilize is a rubric by which to evaluate homework content.21 Since
I want to look for correlations and breakdowns between teacher perceptions of homework
content and actual homework content, I need a way to evaluate the actual homework
content. One might wonder why I think my evaluation of actual homework content is
more objective than the teacher perception of homework ascertained by the survey. First
of all, the teachers’ perceptions ascertained on the survey are done without looking at the
homework (some teachers had not looked at their homework sets for several semesters).
Ideally, I want to evaluate the homework sets in such a way that if the teacher were to
evaluate the homework with me, he would agree with my ratings. I developed a rubric by
which to objectively evaluate the homework for every teacher. The following table is the
rubric for the expectation dimension of concepts. The number on the left is the degree to
which the question is conceptual. 1 is the lowest and 5 is the highest.
Homework Concept Rubric: 21 See Appendix A3 for a copy of the rubric used.
25
1 HW question can be solved by pattern-matching or manipulating one equation.2 HW question can be solved by manipulating fewer than three equations.3 Some knowledge of what symbol(s) in equation represent is helpful in solving question.4 Knowledge of what symbol(s) in equation represent is required to solve question.5 HW question explicitly asks students about concepts, must be answered in terms of concepts
First I acquire homework sets from every teacher I survey and interview. I then
select three representative sets: the second assignment, the fifth or sixth assignment and
the tenth or twelfth assignment. By choosing these assignments I don’t have to worry
about a ‘pretest’ or ‘concluding’ assignment which may skew the evaluations. Then I
look at each problem and evaluate them along the same expectation dimensions I ask
about in the survey.
3. Data: Teacher Surveys and Interviews
Now I am ready to use these developed tools to collect data. The data from the
preliminary survey are as follows:
Statement # 4 5 6 7 8 9 10 11 12 13 14 15
Category E I C R I N N C I C E R
Teacher A 4 5 5 5 5 5 4 3 4 2 3 4Teacher B 4 4 3 4 4 4 2 3 4 4 4 2Teacher C 4 3 3 4 4 5 4 4 4 5 4 3Teacher D 4 2 2 4 3 3 2 3 4 2 3 2
E = Effort C = Coherence N = ConceptsI = Importance R = Reality
1 Strongly Disagree 3 Neutral 5 Strongly Agree
Notice that this survey asks about effort, which is absent from the next survey.
This is because it would be too difficult to try to evaluate the homework problems for
whether students were constructing their own ideas or taking information as given. This
Figure 10: Preliminary Data for Survey #1, by teacher and by category
Figure 9: Rubric designed to evaluate homework sets along the conceptexpectation dimension
26
kind of dimension would be much easier to evaluate in the context of a lecture or lab.
Also, this first survey does not differentiate between the two different kinds of reality
links discussed earlier. The raw data from the revised survey are as follows:
Category: Imp Coh RW Coh Imp SW Con Coh Imp SW Con RW
Statement # 4 5 6 7 8 9 10 11 12 13 14 15
Teacher A 5 5 5 2 5 4 1 3 5 5 5 5
Teacher B 4 4 3 2 5 2 2 4 3 2 4 5
Teacher C 4 4 4 2 3 4 2 4 4 4 5 5
Teacher D 2 4 4 2 3 2 4 3 4 3 3 4
Teacher E 5 4 2 4 5 1 5 2 5 1 2 5
Imp = Importance RW = Real World Con = ConceptsCoh = Coherence SW = Student World
The first question on the survey is for the teacher to list his top five goals for his
class. The table below shows sample data from one of the teachers interviewed.
Complete data from the preliminary questions about the teacher’s goals for the class are
listed in Appendix B3.
Goals Teacher C
1 Think scientifically
2 Develop new ways of "seeing" the universe
3 Develop qualitative and quantitative reasoning
4 Develop skills of explaining your thoughts
5 Critical thinking
The data from the same question on the second survey are listed in Appendix C3.
The table below shows the same teacher’s answer to the same questions one semester
Figure 12: Preliminary Data for Survey #1 Goals for sample Teacher C.
Figure 11: Preliminary Data for Survey #2, by teacher and by category
27
later. Notice that Teacher C does not list content or physics as a goal on the first survey
but does in the second. Despite these changes, his answers are very similar.
Goals Teacher C1 Learning Environment
2 Physical reasoning3 Explain to peers4 Quantitative Reasoning5 Physics
Although most goals were found in more than one teacher’s list, this goal
appeared only once, as goal #5 of Teacher E:
Goals Teacher E5 A necessary filter
The second question on the survey asks the teachers to estimate how much of
student learning occurs via different media. The data from the first teacher interviews for
sample Teacher A and B (complete data in Appendix B4) about what percentage of
learning occurs in these areas are as follows:
% OF LEARNING OCCURS Labs & lectures homework text book exam prep recitationsTeacher A 15% 75% 5% 5% N/ATeacher B 30% 20% 10% 10% 30%
The data from the second teacher interviews about the same question for the same
teachers are as follows: (complete data in Appendix C4)
% OF LEARNING OCCURS labs & lectures homework text book exam prep recitations otherTeacher A 25% 60% 2% 8% 5% 0%Teacher B 25% 30% 12.5% 12.5% 20% 0%
Figure 15: Data for Survey #1 Percentage of Learning for Teachers A and B.
Figure 14: Data for Survey #2 Goals for sample Teacher E.
Figure 13: Preliminary Data for Survey #2 Goals for sample Teacher C.
28
4. Analysis: Teacher Surveys and Interviews
In the previous section about the tools I use (survey and interviews) I explain how
the evaluation of the interviews helps me to make changes to the survey. Happily, I did
not get very many complaints about my second survey. Teachers generally understood
the statements on the survey and felt comfortable agreeing or disagreeing with them.
The notable exception to this is statement #15: “Most of the ‘real world’
homework problems I assign are ‘idealized’ or ‘abstracted’ to simplify the complexities
of the real world.” This statement was intended to be a negative statement about the real
world link. In other words, I was hoping that teachers who believed that their homework
problems were about the real world would answer with a disagreement. Unfortunately,
this statement cued a methodological answer rather than a content answer. What I mean
is that teachers viewed this statement as about the methods by which students solved the
problems. They recognized that many of their homework problems were designed to
teach students how to use abstraction in order to solve complex problems. Their
homework problems still may have had real world content, such as a bicycle, instead of
abstracted content, such as a generic “object”. This confusion clearly presents itself in
my interviews. Teacher C, when answering statement #15 about whether his homework
was abstracted, said, “Absolutely. That’s what physics is all about.”
The misunderstanding about statement #15 is corroborated by the data themselves
as I will show below.
Figure 16: Data for Survey #2 Percentage of Learning for Teachers A and B.
29
Reversing Data
In the second survey, statements 7, 10 and 15 are ‘negative’ statements (where an
agreement constitutes a disagreement with the expert expectation). Therefore, the data
must be reversed for these statements (by exchanging 1 and 5 as well as 2 and 4).
Arranging the reversed data for survey #2, by category, we have:
Category Imp Coh RW SW Con
Statement 4 8 12 5 7 11 6 15 9 13 10 14
Teacher A 5 5 5 5 4 3 5 1 4 5 5 5Teacher B 4 5 3 4 4 4 3 1 2 2 4 4Teacher C 4 3 4 4 4 4 4 1 4 4 4 5Teacher D 3 3 4 4 4 3 4 2 2 3 2 3Teacher E 5 5 5 4 2 2 2 1 1 1 1 2
Graphically, by category, the data are below. The first graph is ‘importance’ and
contains data about how important homework is. The numbers (1-5) on the left of the
graph represent the possible agreement (5 = strong agreement, etc.) The letters along the
bottom represent the different teachers (A – E).
Importance:
1
2
3
4
5
A B C D E
#1 #2 #3
The next four graphs show the data on the four different expectation dimensions
my research project analyzed. If the graph has two bars, it means that the survey asked
the teacher to agree or disagree with two statements about that category. Similarly, if it
Figure 18: Graph of Importance of Homework for three questions for each teacher
Figure 17: Reversed Data for Survey #2 to account for ‘negative statements.’
30
has three bars, there were three statements with which the teacher was asked to agree or
disagree. Again, the numbers on the left represent the level of agreement (1 – 5) while
the letters across the bottom represent the different teachers surveyed.
Coherence: Concepts:
1
2
3
4
5
A B C D E
#1 #2 #3
1
2
3
4
5
A B C D E
#1 #2
World Reality: Student Reality:
1
2
3
4
5
A B C D E
#1 #2
1
2
3
4
5
A B C D E
#1 #2
The data in these graphs offer another check for the validity of my survey. First,
there should be general agreement for the statements of any given category. For
example, a teacher who strongly agrees with the statement: “If a student does poorly on
the homework in the first unit, he or she can still do well on the next unit’s homework by
working exclusively on the new material” should also strongly agree with the statement:
Figure 19: Graphs of teacher perception of 4 Expectation Dimensions in Homework
31
“There is usually at least one homework question in each set about how the current
material relates to other material from the course.” If questions from one homework set
ask about how the material relates to other material, a student cannot focus exclusively on
the new material and do well. If, on the other hand, there is significant disagreement, it is
probably because the teacher doesn’t think the statement is about what I intended it to be
about.
This is, in fact, what happened for the statements about the real world link. The
graph for the world reality link is as follows:
1
2
3
4
5
A B C D E
#1 #2
Clearly, teachers think the two statements are asking something very different.
Most teachers strongly disagreed with the second statement while agreeing, in some
cases, with the first. It is important to note that there is no correlation for any teacher
between bar #1 and #2. Therefore, I know that the two statements
• Homework problems frequently ask the student to consider a realworld situation to which a physics principle applies.
and• Most of the ‘real world’ homework problems I assign are ‘idealized’
or ‘abstracted’ to simplify the complexities of the real world.
Figure 20: Graph of teacher perception of Real World Expectation Dimension in HW
32
are not about the same expectation. As I discussed above, the second statement is
intended to be a negative statement about the real world. A negative answer is taken to
mean that physics problems are about the real world. However, as I noted above, most
teachers said that all physics problems are abstracted, and furthermore, that’s how it
ought to be. Therefore, I discard the second statement and use only the first statement to
determine the teachers’ beliefs about the real world link in their homework. This
category on the survey will not be as robust as if I had two statements, but I can still
obtain data with the other statement about the real world link.
For the other three expectation dimensions, excluding importance (which is used
to ascertain the overall importance of homework rather than beliefs and attitudes
expressed within the homework), I average the results. For example, since teacher B
agreed with a ‘4’ for both questions on concepts, his average value is a ‘4’ for concepts.
5. Data: Homework
Next I took data on the homework sets of the teachers. I did this by selecting
three representative homework sets. Then I evaluated every question of these homework
sets for its real world link, its student world link and its concepts. Then I evaluated each
of the three homework sets for its overall coherence. I evaluated them according to the
homework rubric I developed in collaboration with Professor Finkelstein.22
I will demonstrate the method I used for evaluating the homework for one sample
class: Teacher B’s class. I used the same method for evaluating the homework in every
class, so I will skip the method and report the results for the other classes.
22 See Appendix A3
33
First, I selected three representative homework sets. For example: #2, #6, and
#11. Then, I evaluated every question of these three homework sets for Real World Link,
Student World Link and Concepts according to the rubric discussed above:
HW2 WR SR CON#1 3 3 2
#2 3 3 3#3 3 4 3#4 3 3 2#5 2 2 5
#6 2 2 3#7 1 1 2#8 1 1 3
HW 6 WR SR CON#1 3 4 3
#2 3 3 5#3 2 1 3#4 4 4 3#5 3 3 3
#6 4 4 4
HW 11 WR SR CON#1 3 3 3
#2 2 2 4#3 2 2 3#4 2 2 4#5 3 3 3
#6 4 3 4#7 3 3 5
Then I evaluated each set for coherence:
CoherenceHW 2: 3.00HW 6: 3.00
HW 11: 4.00
The evaluations for each of the other homework questions for the different
teachers can be viewed in Appendix D5.
6. Analysis: Homework
After I had these values, I averaged the homework values for the four expectation
dimensions. For example, for calculating the average for student reality, I added up all of
the values and divided by the total number of homework questions. I did not average in
stages; first by questions in the homework set, then between the sets themselves. This is
because each homework question is worth the same number of points. Assignment #2
has eight questions and is therefore worth eight points, whereas assignment #6 has six
Figure 21: Table of Coherence Expectation Dimensions in each 3 sets of HW
Figure 21: Table of three Expectation Dimensions in each problem of 3 sets of HW
34
questions and is therefore only worth six points. Since each question has the same
weight, it should be averaged with equal weight. Therefore, we are left with the
following table of average values:
Homework SummaryTeacher A B C D FWR 4.73 2.67 2.76 2.9 2.48SR 4.4 2.67 2.44 2.05 1.52CON 4.73 3.33 2.92 2.85 2COH 4.33 3.33 3.67 4 3.67
Graphically, we have the same homework summary:
1
2
3
4
5
A B C D F
WR
SR
CON
COH
Discussiona. Comparative Analysis: Teacher Perceptions and Homework
Now, with these values, I am ready to proceed to an analysis where I compare the
teachers’ perceptions of the actual homework content with evaluations of the actual
homework content along these expectation dimensions and look for evidence of any
discontinuities. The first bar in the following graphs represents the actual homework
content and the second bar represents the teachers’ perception of their homework.
Figure 23: Graph Summarizing 4 expectation dimensions in HW for each teacher
Figure 22: Table Summarizing 4 expectation dimensions in HW for each teacher
35
TEACHER A: TEACHER B:
1
2
3
4
5
WR SR CON COH
Actual HW Content Percieved HW Content
1
2
3
4
5
WR SR CON COH
Actual HW Content Percieved HW Content
TEACHER C: TEACHER D:
1
2
3
4
5
WR SR CON COH
Actual HW Content Percieved HW Content
1
2
3
4
5
WR SR CON COH
Actual HW Content Percieved HW Content
Teacher E:
1
2
3
4
5
WR SR CON COH
Actual HW Content Percieved HW Content
36
The first discrepancy I will look for is between teachers’ perception of
expectations in homework and actual expectations in homework. Teacher A has the most
consistency between his perceptions of his HW and the HW itself. The most the two
values disagree is .27, less than a third of a point. The second most consistent teacher is
Teacher B whose largest discrepancy is .67, two thirds of a point.
Next, we have Teacher D and Teacher E. Both of these teachers have strong
agreement for three out of four categories, but a sizeable discrepancy between their
perceptions of their homework and their homework content in one dimension. For
Teacher D, it is the World Reality Link and for Teacher E, it is Coherence. Not only are
these large discrepancies (roughly 1 whole point), but the two values are split across #3.
Remember that 3 is neutral, so above a three is an agreement and below, a disagreement.
Therefore Teacher D agreed that his homework is about the real world, but the evaluation
of the homework disagrees. Similarly, Teacher E disagreed that his homework was
coherent, but the evaluation of the homework agrees with the statement that the
homework is coherent.
Teacher C, the least consistent, has a large discrepancy for three out of four
expectation dimensions: World Reality, Student Reality and Concepts. This teacher
perceived his homework to contain much more of the expectations than it actually did.
He overestimated these three categories by 1.24, 1.56, and 1.58, respectively. This
discrepancy means that he agreed that his homework was about the real world, about his
students’ world and required conceptual understanding. However, his homework, when
evaluated, does not contain as many real world, student world and conceptual problems
as he believed.
37
b. Comparative Analysis: Teacher Perceptions and Teacher Goals
For the teachers whose perceptions of their homework were consistent with the
researchers’ analysis, it may be fruitful to also look at their goals for consistencies and
discrepancies with their perceptions of their homework. For instance, Teacher E has
consistent perceptions with homework content for world reality, student reality and
concepts; each of which is low (between 1 and 2.5). There are a few possible
explanations for this. First, if he holds these expectations as important goals of the class,
then either he does not think homework is the place to teach these ideas, or he does think
that homework is the place to teach the ideas. If the latter is the case, then we have
discovered a breakdown between his ideal physics homework content and his perception
of the homework content. This is clearly the case for Teacher B who has consistent
perceptions of homework with homework content, said 30% of student learning occurred
while doing homework, and who exclaimed during the interview that he did all of these
reforms in his lectures and tutorials, but, “no, I don’t do any of this in the homework.”
His frustration indicated his awareness of the breakdown between his ideals and his
perceptions.
Returning to Teacher E, recall that he has consistent but low values for
perceptions and content of homework. Teacher E’s goals are the following:
GOALS Teacher E1 Confidence in the force of reasons2 skills and career paths
3 create good citizens -informed and inquisitive4 grasp of concepts5 a necessary filter
Figure 24: Table of Goals for Teacher E
38
Therefore, if this teacher is concerned with career skills, creating good citizens
and providing a filter, he may not be as concerned with students learning (the
expectation) that physics is a conceptual field about their world and the real world.
Interestingly, his fourth goal is “a grasp of the concepts”. This can mean one of two
things: either a breakdown exists between his ideals and his perceptions, or he means
something different by ‘concept’ than do others.
There is another possibility for Teacher E that I will mention here. Teacher E
thinks that only 15% of student learning occurs during homework. Therefore, he may not
emphasize expectations he wants his students to learn in the homework because he
believes that more student learning happens in other places.
Teacher A, as a comparison, has extremely high consistency between his
perceptions of his homework and his homework content. To compare these results to his
goals, we have:
GOALS Teacher A1 World around them understand and explain by physics2 reasoning ability applying principles
3 quantitative problem solving and understand its value4 appreciate and value physics5 understand and apply specific content
Clearly, Teacher A believes that ideally, students will learn about the world
around them (goal #1). Teacher A had very high values for world and student reality
links (4.73, 4.4) in the homework and believed that those were very high (5, 4.5).
Therefore, we should see that Teacher A believes homework is an important media since
he has changed it to contain such a high amount of the real and student world
Figure 25: Table of Goals for Teacher A
39
expectations. In fact, this is exactly what we see. Teacher A thinks that 60% of student
learning occurs while doing homework. Therefore, he not only has consistency between
his perceptions of homework and his homework, he also has consistency between his
ideal homework content and his perception of homework content. Therefore, in my
model:
The breakdown for Teacher A does not happen at the first or second question
mark. Therefore, if the students are not completing the class with more favorable
expectations than they had before the class, then the breakdown will be happening
somewhere else. There is a slight breakdown at the second question mark for Teacher B,
but the largest breakdown for him is at the first question mark between what he wants to
teach in the homework and what he thinks is in the homework. For Teachers D and E,
there is a non-trivial breakdown at the second question mark. At the first question mark,
Teacher D has some breakdown while Teacher E has very little. Teacher C has a large
breakdown at the second question mark between his perceptions of the homework and the
actual homework content.
Figure 8: Model of interaction between teachers and students within homework
Teachers’ Perception of Actual Homework Content
Teachers’ Ideas of Physics and Ideal Homework Content
Actual Homework Content
Students’ Perception of Actual Homework Content
Students’ Ideas of Physics
40
c. Comparative Analysis: Familiarity with Physics Education Research
It is interesting to look at which of Teachers A-E has studied or is familiar with
physics education research. Based on the teachers’ self-reported familiarity as well as the
interviews, I was able to characterize the five teachers according to their familiarity with
PER. Teacher A is highly aware of physics education research and is actively engaged
in research-based reform. Teacher B is also highly aware of physics education research
and is engaged, though not as actively as Teacher A in reform. These teachers can be
categorized as “Fully” aware of PER. Teachers C and D both have some knowledge of
the field of PER and are moderately engaged in some reform. They can be categorized
as, “Moderately” aware of PER. Teacher E has just begun looking at results from PER
and is interested in beginning some moderate reforms. He can be categorized as “Lowly”
aware of PER.
Significantly, the teaching practices of these teachers can also be characterized
along a parallel axis. Teachers A and B are “Reform” teachers, actively engaging in
research-based changes to their classes. Teachers C and D are “Transitional” teachers
who have made some significant changes to their classes. Teacher E teaches traditionally
and has made very few changes to his class. We can categorize him as “Traditional”.
Teacher Awareness of PERExtent of Reform inClass
A Fully Aware of PER ReformB Fully Aware of PER ReformC Moderately Aware of PER TransitionalD Moderately Aware of PER TransitionalE Lowly Aware of PER Traditional
Figure 26: Table of Characterizations of Teachers
41
These preliminary characterizations allow us to notice that the two most
consistent teachers, as measured by accurate perception of homework, are fully aware of
PER and actively engaged in sizeable reform of their classes. The next most consistent
teachers were the traditional and the second transitional teachers. The least consistent
teacher was moderately aware of PER and was transitional. This could indicate a variety
of causal factors which my study is unable to differentiate between. For instance,
transitional teachers may be less consistent because their goals change more rapidly than
their practices. Or, they may overestimate the extent to which expectations are present,
still implicitly believing that expectations ‘come along for the ride’ with content. These
are very interesting questions and would be worth pursuing.
IV. Future Work
There are a lot of areas for further research. It would be very interesting to
correlate these findings with data on student expectations. For instance, do students of
Teacher A have higher or lower expectations than those of Teacher C? Than Teacher E?
That research would complete the vertical analysis according to my theoretical model
(fig. 8). To extend the research horizontally in the same model, it would be interesting to
look at teacher perceptions of labs, lectures, exams, etc., then compare those perceptions
with independent evaluations of these media. It would also be very interesting to
quantify the awareness of PER and teaching styles more quantitatively. I presented just a
very basic outline of what it might look like in the previous section.
Another area for interesting future work is to acquire more statistically robust
data. Now that I have validated the second survey (with the exception of question #15)
the survey could be more widely distributed to teachers without a corresponding
42
interview. This way, a much larger amount of data could be collected and correlated to
actual homework content.
V. Conclusion
My honors thesis was to pick a problem in physics education research, research it
and acquire and analyze data. I chose to examine teacher goals, teacher beliefs about
expectations in their homework as well as the actual expectations in their homework. In
this paper I have presented background material in Physics Education Research. I have
motivated the problems of concepts and expectations not getting taught to intro-level
undergraduates. I then presented my project. I presented the conceptual space and
theoretical models with which I was working. I demonstrated the tools I developed to
inquire into this topic. I developed a survey, conducted interviews and evaluated
homework sets according to the rubric I largely designed. I acquired data and analyzed
it. Overall, it was a terrific experience because I was able to pick the problem I wanted to
look into and I was able to design the most of the tools I used to inquire into these issues.
I obtained many interesting conclusions about characterizing where breakdown occurs
between teachers and their students and their homework.
43
Appendix A1: Survey #1
Questions
1. Please list (in order) your top five goals for this class.a. b. c. d. e.
2. Out of the following, what percentage of learning occurs:a.) During lectures b.) While doing homework c.) While reading text book d.) While preparing for exams e.) During recitations / labs f.) Other
3. I write or design ______% of homework problems assigned in my class.
Please answer the following questions according to the scale:1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree
4. If students do not do well on homework assignments it is generally because theyare not putting in as much time and effort as I have recommended.
5. Everything I expect my students to know on an exam is covered at some point onthe homework.
6. The overall coherence of physics is often explicit in homework.7. Homework problems frequently ask the student to consider a real world situation
to which a physics principle applies.8. If a student does poorly on the homework, he or she will most likely do poorly
overall in the class.9. Most homework problems can be solved by carefully looking over the given
equations to find the right one.10. Students are asked to infer general principles from specific examples in the
homework problem sets.11. If a student does poorly on the homework in the first unit, he or she can still do
fine in the next unit by working harder exclusively on the new material.12. Students know what I think it is important for them to learn because of the
homework problems I assign.13. There are multiple approaches or ways to solve the homework problems I assign.
44
14. If a student is having trouble, I encourage him or her to try harder on thehomework.
15. Homework questions often ask the student to consider his or her own experience.
Appendix A2: Survey #2
Questions1. Please list (in order) your top five goals for this class.
a. b. c. d. e.
2. Out of the following, what percentage of learning occurs:a. During lectures b. While doing homework c. While reading text book d. While preparing for exams e. During recitations / labs f. Other
Please answer the following questions according to the scale:1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree
3. I am familiar with physics education research that has been done on coursessimilar to mine.
4. All of the material (concepts, formulas, etc.) I expect my students to know on anexam is covered at some point on the homework.
5. The majority of problems in a given homework set ask about the same two orthree ideas.
6. Homework problems frequently ask the student to consider a real world situationto which a physics principle applies.
7. If a student does poorly on the homework in the first unit, he or she can still dowell on the next unit’s homework by working exclusively on the new material.
8. If a student does poorly on the homework, he or she will most likely do poorlyoverall in the class.
9. Homework questions often ask the student to consider his or her own experience.10. Most homework problems can be solved by carefully looking over the given
equations to find the right one.11. There is usually at least one homework question in each set about how the current
material relates to other material from the course.12. If students want to know what I think it is important for them to learn, they can
look at the homework problems I assign.13. The material I cover in my homework relates to the students’ everyday lives.14. In order to solve my homework problems students must understand the
underlying concepts.
45
15. Most of the ‘real world’ homework problems I assign are ‘idealized’ or‘abstracted’ to simplify the complexities of the real world.
Please use this space to write any crucial information left out of the survey.
Appendix A3: Homework Evaluation \Rubric
Grading Rubric:Coherence1 HW set contains six or more different ideas or concepts2 HW set contains five different ideas or concepts3 HW set contains four different ideas or concepts4 HW set contains three different ideas or concepts5 HW set contains one or two different ideas or concepts
(Add one to this score if at least one question explicitly relates this material to previousmaterial.)(Subtract one from this score if no question explicitly relates this material to previous material.)
Concepts1 HW question can be solved by pattern-matching or manipulating one equation.2 HW question can be solved by manipulating fewer than three equations.3 Some knowledge of what symbol(s) in equation represent is helpful in solving question.4 Knowledge of what symbol(s) in equation represent is required to solve question.5 HW question explicitly asks students about concepts, must be answered in terms of concepts
World Reality1 No link to real world, question is purely abstracted2 Has recognizable features that could be construed as real world3 Real world features are used, but abstraction is expected4 Real world features are used, but abstraction is sufficient to answer question
5Real world features and objects are used; knowledge of real world is required to answerquestion
Student Reality1 No link to student's life2 Has recognizable features that could be construed in terms of student's life3 Features from student's life are used, but abstraction is expected4 Features from student's life are used, but abstraction is sufficient5 Features from student's life are required to answer question
46
Appendix B1: Survey #1 Data
Statement # 4 5 6 7 8 9 10 11 12 13 14 15
Category E I C R I N N C I C E R
Teacher A 4 5 5 5 5 5 4 3 4 2 3 4Teacher B 4 4 3 4 4 4 2 3 4 4 4 2Teacher C 4 3 3 4 4 5 4 4 4 5 4 3Teacher D 4 2 2 4 3 3 2 3 4 2 3 2
Appendix B2: Reversed Survey #1 Data
Person E E I I I C C C R R N NTeacherA 4 3 5 5 4 5 3 2 5 4 5 4TeacherC 4 3 2 3 4 2 3 2 4 2 3 2TeacherD 4 4 3 4 4 3 4 5 4 3 5 4TeacherB 4 4 4 4 4 3 3 4 4 2 4 2
Appendix B3: Interview #1 Data: Goals:
Goals Teacher A Goals Teacher C
1 Be able to physics in everyday life 1 Think scientifically
2 "Expert" attitudes about physics 2 Develop new ways of "seeing" the universe
3 Logical problem solving approaches 3 Develop qualitative and quantitative reasoning
4 Understand numbers can be useful 4 Develop skills of explaining your thoughts
5 Gain appreciation of physics 5 Critical thinking
Goals Teacher B Goals Teacher D
1 Content (conservation laws) 1 Importance of Physics
2 Qualitative understanding 2 Learn how basic things work
3 Articulate / Evaluate 3 Problem solving skills
4 Coherence 4 Understand how science develops
47
5 Enjoy physics 5
Appendix B4: Interview #1 Data: % of Learning Occurs:
lectures homework text book exam prep recitationsTeacher A 15% 75% 5% 5% N/ATeacher B 30% 20% 10% 10% 30%Teacher C 20% 30% 10% 20% 20%Teacher D 30% 30% 20% 10% 10%
Appendix C1: Survey #2 Data:
Appendix C2: Reversed Survey #2 Data
(Negative Agreement has been Reversed)
Category: PER Imp Coh RW Coh Imp SW Con Coh Imp SW Con RW
Question # 3 4 5 6 7 8 9 10 11 12 13 14 15
Teacher A 5 5 5 5 4 5 4 5 3 5 5 5 1
Teacher B 5 4 4 3 4 5 2 4 4 3 2 4 1
Teacher C 4 4 4 4 4 3 4 4 4 4 4 5 1
Teacher D 4 2 4 4 4 3 2 2 3 4 3 3 2
Teacher F 4 5 4 2 2 5 1 1 2 5 1 2 1
Category: PER Imp Coh RW Coh Imp SW Con Coh Imp SW Con RW
Question # 3 4 5 6 7 8 9 10 11 12 13 14 15
Teacher A 5 5 5 5 2 5 4 1 3 5 5 5 5
Teacher B 5 4 4 3 2 5 2 2 4 3 2 4 5
Teacher C 4 4 4 4 2 3 4 2 4 4 4 5 5
Teacher D 4 2 4 4 2 3 2 4 3 4 3 3 4
Teacher F 4 5 4 2 4 5 1 5 2 5 1 2 5
48
Appendix C3: Interview Data: Goals
GOALS Teacher A1 World around them understand and explain by physics2 reasoning ability applying principles3 quantitative problem solving and understand its value
4 appreciate and value physics5 understand and apply specific content
GOALS Teacher B
1 Content Knowledge -conceptual, factual, formal2 Problem solving skills3 Improved ability to "talk physics" -includes scientific argument
4 Reality Link -physics is about the world5 Self checking: improved ability to know what they know, sense making
GOALS Teacher C1 Learning Environment
2 Physical reasoning3 Explain to peers4 Quantitative Reasoning
5 PhysicsGOALS Teacher D
1 Appreciate importance of physics in society2 how using abstract concepts describes real events
3 appreciation on scientific method4 ability to understand basic E&M (e.g. how a microwave works)5
GOALS Teacher F1 Confidence in the force of reasons2 skills and career paths3 create good citizens -informed and inquisitive
4 grasp of concepts5 a necessary filter
49
Appendix C4: Interview Data: Importance of Media
% OF LEARNING OCCURS labs &
lectures Homework text book exam prep recitations other
Teacher A 25% 60% 2% 8% 5% 0%
Teacher B 25% 30% 12.5% 12.5% 20% 0%
Teacher C 30% 30% 10% 10% 20% 0%
Teacher D 20% 20% 20% 20% 3% 17%
Teacher E 25% 15% 10% 30% 20% 0%
Appendix D5: Homework Data
TeacherB WR SR CON COH
HW set 2 3 3 2
3 3 3
3 4 3
3 3 2
2 2 5
2 2 3
1 1 2
1 1 3 3
HW set 6 3 4 3
3 3 5
2 1 3
4 4 3
3 3 3
4 4 4 3HW set11 3 3 3
2 2 4
2 2 3
2 2 4
3 3 3
4 3 4
Teacher A WR SR CON COH
HW set 2 5 5 5
5 5 5
5 5 5
5 5 5
5 5 5 4
HW set 7 5 3 5
4 4 5
4 4 4
4 4 4 4
HW set 12 5 4 5
5 5 5
5 4 5
4 3 3
5 5 5
5 5 5 5
Averages: 4.73 4.4 4.73 4.33
50
3 3 5 4
Averages: 2.67 2.67 3.33 3.33
TeacherC WR SR CON COH
HW set 2 2 3 4
2 2 3
2 2 3
2 2 3
2 2 3
2 2 4
2 2 2
2 2 3
3 3 4 4
HW set 6 4 3 2
3 3 2
3 2 2
3 2 3
3 2 2
3 2 2 4HW set11 4 2 1
3 3 3
3 3 4
3 3 4
3 2 3
2 2 2
2 2 3
3 2 3
4 4 4
4 4 4 3
Averages: 2.76 2.44 2.92 3.67
TeacherD WR SR CON COH
HW set 2 3 2 3
2 2 3
2 1 1
2 1 3
3 1 5
3 1 5
3 2 4 4
HW set 6 2 2 1
3 1 2
3 2 3
3 2 2
2 2 4 4HW set11 3 1 1
3 3 2
4 3 4
4 3 4
3 4 3
3 2 1
3 3 3
4 3 3 4
Averages: 2.9 2.05 2.85 4
51
TeacherE WR SR CON COHHW set 2 3 1 1 1 1 1 2 1 2 2 2 2 2 1 1 2 1 3 2 1 2 2 1 2 3 2 2 3.00HW set 7 3 2 1 2 1 2 4 2 3 3 1 2 2 1 2 3 1 2 2 1 2 2 1 3 2 1 2 4.00 3 3 3 3 2 2 4 3 2 3 2 2 2 2 1 2 1 2 3 2 2 2 1 2 3 3 3 4.00Averages: 2.48 1.52 2 3.67
52
Appendix C1: Resources: MPEX
53
54
Bibliography
Physics Education Technology, Colorado Learning Attitudes about Science Surveyhttp://consmos.colorado.edu/phet/survey /fall 2003.
Halloun, “Views about science and physics achievement: The VASS story,” AIP Conf.Proceedings 399 (1997).
Hammer, David, “Student Resources for learning introductory physics,” AmericanJournal of Physic, Physics Education Research Supplement, 68 (S1), S52-S59.
Hammer, David, “Epistemological beliefs in introductory physics,” Cognition andInstruction 12(2), 151-183 (1994).
Henderson, Charles, “Easier Said than Done: A Case Study of Instructional ChangeUnder the Best of Circumstances,” Western Michigan University, Kalamazoo, MI.
Henderson, Charles, “Grading Student Problem Solutions: The Challenge of Sending aConsistent Message,” Western Michigan University, Kalamazoo, MI.
Hestenes, Wells, and Swackhamer, “Force Concept Inventory,” Physics Teacher 30 141-158 (1992).
McDermott, L.C., “Guest comment: How we teach and how students learn –amismatch?” American Journal of Physics, 61, 295-298 (1993).
Redish, Edward, Teaching Physics, John Wiley and Sons, 2003.
Redish, Saul and Steinberg, “Student Expectations in introductory physics,”MarylandPhysics Expectations Survey University of Maryland Physics Education Research Group.American Journal of Physics 66 212-224, (1998).http://www.physics.umd.edu/perg/expects/mpex.htm.
TIMSS: Mathematics and Science Achievement in the Final Year of SecondarySchool. February 1998. Mullis, Martin, Beaton, Gonzalez,Kelly, and Smith(http://timss.bc.edu/timss1995i/TOC_MSC.html)